JP2003167805A - Network communication method among two or more user- side closed networks and server-side closed network, and server device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To operate a physical one server device of a server-side closed network as an exclusive server by not sharing the same among user-side closed networks, but independently providing the same to each of user-side closed networks. <P>SOLUTION: Two or more user-side closed networks 31-33 are connected to the server-side closed network 10 by access servers 26, 28-30 by using an IP tunnel. When the connection is required, the access server 26 discriminates the closed network having made the requirement, attaches a specific identification code to the user-side closed network by means of a packet tag, and transfers the same to the server device 11. The server device 11 comprises multiple upper communication flows by protocol stacks 19-21, and lower communication flows from the protocol stacks 19-21 to logical device interfaces 23-25, and these communication flows 12-14 are assigned to the user-side closed networks 31-33. The server device 11 specifies the closed network of a transmitting source host on the basis of the identification tag of the received packet. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a network communication method such as an application hosting service for an extranet and an information sharing server service for communicating between a plurality of different closed networks connected via the Internet or a private network, and Regarding the device, for example, a server service that physically provides a user with a closed network for a plurality of users by using one server, or a service that shares files and databases with a plurality of closed networks through a server. The present invention relates to a network communication method and a server device between a closed network on the side of a plurality of users and a closed network on the side of a server.

[0002]

2. Description of the Related Art The use of network communication such as the Internet is widely spread, and companies and communities use closed networks to provide intranets and extranets in order to keep communication costs low and to perform secure communication. I'm building. In recent years, closed networks are not limited to large enterprises, but also small and medium enterprises and SOHO (Small Office H
ome Office), and the number tends to increase. Furthermore, LAN (L
The usage form of the closed network is diversifying, such as the construction of a LAN in an ocal area network (for example, a form in which a hierarchical net is built in an intranet).

As a method of providing a server service to such a small-scale and large number of closed networks, there is a method of installing a server in each closed network, but the cost and operation are high and the burden on the server installation side is heavy. Therefore, although one server is physically used to provide a server service to each closed network, each user's closed network does not have to share the Internet communication function and the server process between the user's closed networks. We will consider the construction of a virtual private server that enables independent networks and operates as if each closed network had its own dedicated server.

In the construction of a closed network, it takes a lot of time and cost to obtain a global address, so that a method using a local address is often adopted. Considering the provision of a server service by physically using one server for a plurality of closed networks constructed in this way, the following problems must be solved. (1) Collision of local address space between closed networks including server-side closed networks (2) Unauthorized access such as crossing other closed networks via server (3) From server to host in each closed network (4) Ease of adding a closed network including the server side and changing the network configuration

As a conventional technique for communicating between a plurality of user-side closed networks and a server-side closed network, L2TP is used.
(Layer 2 Tunneling Protocol
ol) and IPsec (IP security pro)
IP (Internet) such as tocol)
Protocol) tunnel technology, NAT (Network)
ork Address Translation and R-NAT (see Japanese Patent Application No. 2001-016255).
Root side-Network Address
There are three techniques, an address conversion technique typified by Translation, and a technique using both the IP tunnel technique and the address translation technique.

IP, which is one of the IP tunnel technologies,
The tunnel mode of sec is a virtual connection that securely connects between closed networks by adding an IP header used for transfer in the wide area network to an IP packet from the closed network, encapsulating it, and performing encryption processing. This is a technique for creating a tunnel. In L2TP, a terminal connects to an access server installed in an internet provider or the like via a public telephone network, and this server establishes a virtual tunnel with a server prepared in the closed network, thereby Between the closed network and the PPP (Point
It is as if a connection by t-to-Point Protocol) was established. Also, NAT is
It defines a method of converting a private IP address used on the LAN side to a global IP address assigned on the Internet side when connecting the LAN to the Internet.

Now, it is assumed that the user-side closed networks A and B are connected to the server-side closed network S via the Internet by an inter-network access server using an IP tunnel. Hosts and S belonging to the user-side closed networks A and B
Address spaces A, B, and S are assigned to the servers belonging to. With this configuration, each host can be made to appear to be connected to a server belonging to the same closed network. As a result, the local address assigned to each closed network can be used as it is. However, in the closed network, local addresses are freely assigned to the host, so that the address space may overlap between the user-side closed networks or between the user-side closed network and the server-side closed network. If the address space overlaps between the user-side closed networks, there are multiple hosts with the same address from the perspective of the server, and the server cannot correctly recognize which host it is communicating with. , The problem arises.

Further, in order for a host belonging to the closed network on the user side and assigned a local address to communicate with a server assigned a global address on the Internet, the host must also have a global address. At this time, a global address is statically assigned to the NAT device provided at the boundary between the closed network and the Internet, and the global address and the local address assigned to the host are associated with each other, thereby enabling the connection between the host and the server. To do.

However, the NAT technique is established by a connection request from the host in order to identify the connection by dynamically associating the local address and the global address with the connection request from the host to which the local address is assigned. Once the connection between the host and the server being disconnected is lost, the correspondence between the local address used for the connection and the global address becomes invalid, so the local address is statically set from the server to which the global address is assigned. Push-type services such as sending information to assigned hosts are not possible. Further, since the server cannot identify the closed network, it becomes difficult to provide the services and information unique to the closed network to the hosts belonging to each closed network.

Therefore, a plurality of user-side closed networks and server-side closed networks are connected using respective IP tunnels, and I
In the access server of the server-side closed network that terminates the P tunnel, a unique IP address is assigned to the host-IP network pair of the user-side closed network, and this is stored. By associating this IP address with the local address assigned to the host, the connection between the host and the server becomes possible. In this way, since the IP address that is different for each host is used from the address space that specifies the closed network, the address translation is performed, so that the collision of the local address space between the connected user-side closed networks and the server can be avoided. It is possible to connect to a specific host.

However, in each user-side closed network,
It is not permissible to assign addresses to servers using free local addresses as if they belong to their network. A server address that is unique to each user's closed network, taking into consideration the local address space used by the other user's closed network and the local address space that other user's closed networks desire to allocate to the server. The problem arises of having to decide. As a result, an environment is not satisfied in which one server is physically used to operate as if there is a dedicated server in each user-side closed network.

Further, when a certain server application is provided to each user-side closed network, the server often uses a common server process to provide all user-side closed networks. Separation between the closed networks on the user side is performed only by the difference in the address that connects to the server, so it becomes impossible to prevent unauthorized access such as crossing other closed networks on the user side via the server, and at the same time on the server and the server. Since the server application that is executed on the server is always forced to be aware of the difference in address, the number of locations to be set is extremely large, and as a result, it is difficult to add a closed network or change the network configuration. Occurs.

[0013]

Conventionally, in Internet communication, the TCP / IP protocol group has been used for communication between devices such as hosts and routers. The TCP / IP protocol group is called an internet protocol stack. The role of this protocol stack is to hold an IP address (server address) assigned to a physical network interface of a server device or a plurality of virtual logical device interfaces, hold route (routing) information for packet delivery,
It is responsible for network communication functions in the Internet environment such as processing of each protocol packet.

FIG. 2 is a diagram showing a conventional Internet protocol stack. The link layer 63 including the physical layer and the data link layer includes hardware interfaces (including drivers) 632, ARP (Address).
Resolution Protocol (633) and RARP (Reverse Address Re)
solution protocol 631. ARP 633 is the IP address
The RARP 631 is a protocol for requesting the hardware address (MAC address or the like) of the node corresponding to the P address.
This is a protocol for requesting the IP address of the own node based on the C address.

The network layer 62 includes an IP processing unit 62.
2, ICMP (Internet Control M)
ESSAGE PROTOCOL) 623, and IG
MP (Internet Group Management)
ent Protocol) 621. ICMP 623 is a protocol used to manage information regarding the state of the IP protocol.
GMP621 is a protocol for realizing IP multicast.

The transport layer 61 is a TCP (Tr
nsmission Control Protocol
l) 612 and UDP (User Datagram P)
protocol) 611. TCP6
Reference numeral 12 is a higher-level protocol of IP, which provides connection-type and highly reliable communication, and UDP 611 is a connectionless-type protocol, which does not perform processing for ensuring reliability, and therefore can perform processing faster than TCP. The application layer 60 uses the user process 601 to provide a communication service that the end user directly uses.

The conventional server device is provided with only one protocol stack, and the hosts and routers that communicate with this server device are provided with the Internet communication function provided by this protocol stack and this protocol stack. Shared with other server processes. Therefore, in the case of a server device that deploys a server service to a plurality of user-side networks, many of them have to perform the service separately for each user-side network, and therefore the server device specifies and separates the user-side network. is necessary.

In the conventional server technology, the difference between the source address and the destination address of the packet transmitted to the server is used to identify and separate the user side network, but the user side network uses the local address space. In the case of a closed network that can be freely assigned, in order to avoid local address space conflicts between closed networks, forcible address management such as address restriction and translation is performed, and It facilitates identification and separation. However, providing the network communication function of the server device and the server process in a shared manner on each of the user side networks is performed by address management or the like.
It is difficult to realize network services such as closedness, which is a feature of closed networks, push-type services that are voluntary information distribution from servers, and ease of adding closed networks and changing network configurations.

Therefore, an object of the present invention is to solve these conventional problems, and in network communication in which a plurality of user-side closed networks are connected to a server-side closed network by an access server using an IP tunnel, the server-side closed network is used. It is possible to independently provide each user-side closed network without sharing the internet communication function and server process physically provided by one server device installed in each user-side closed network. To provide a network communication method between a plurality of user-side closed networks and a server-side closed network and a server device capable of realizing an operation as if the server device is a dedicated server belonging to each user-side closed network. It is in.

[0020]

In order to achieve the above object, in a network communication method between a closed network of a plurality of users and a closed network of a server according to the present invention, in a server device,
Equipped with multiple communication flows consisting of a combination of upper communication flow from server process group to protocol stack and lower communication flow from protocol stack to physical network interface or logical device interface. This communication flow is provided for each closed network on the user side. 1
By allocating them one by one, the individual communication paths in the server between each user-side closed network and the server are logically separated on the network, and the user-side closed networks that are connected to each other are specified and separated.

The server device of the present invention uses the unique communication flow assigned to each user-side closed network to hold the server address carried by the protocol stack, control the packet delivery route, and process internet packets such as processing each protocol packet. The function and the server process such as the Web server and the FTP server are provided independently, and the communication flow between the user-side closed networks is not shared with other user-side closed networks.
On the network, communication with the server device is realized while logically maintaining closedness.

According to the present invention, in a network communication in which a plurality of user-side closed networks are connected to a server-side closed network by an access server using an IP tunnel, a physically single server device installed in the server-side closed network is used. It is possible to operate as if the server device is a dedicated server belonging to each user-side closed network without sharing the provided Internet communication function and server process between the user-side closed networks. become.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a conceptual diagram of a system for realizing the inter-network communication method of the present invention. (Structure) Here, a structure in the case where the communication flow and the external connection are 3 is shown. Multiple user-side closed networks (A) 3
1, (B) 32, (C) 33 and the server-side closed network 10 access servers 26, 2 using an IP tunnel.
Connection is made by 8, 29, 30. The hosts a, b, and c that belong to each of the user-side closed networks 31 to 33 receive the server service from one physically installed server device 11 installed in the server-side closed network 10. Regarding the hosting service of the access server device 26 of the server-side closed network 10, a recording medium recording the "inter-closed network connection system and the closed inter-network connection method and its processing program" described in Japanese Patent Laid-Open No. 2001-016255. Hosting service system ".

The server device 11 has about 200 communication flows (3, 12, 13, 14 in the figure), and each protocol stack 19, 20, 21 has a conventional Internet as shown in FIG. Same as protocol stack. That is, the server process group 15, 16, 1
7 and the TCP / IP protocol groups 19, 20, 21 are connected via the upper mapping module 18, and the TCP / IP protocol groups 19, 20, 21 and the logical device interfaces 23, 24, 25 are connected to the lower mapping module 22. Connected through. The server device 11 includes logical device interfaces 23, 24, 25.
As many as these protocol stacks.

The server device 11 has protocol stacks 19, 20, 21 linked to the respective logical device interfaces 23, 24, 25 and each protocol stack 1
Server process group 15,1 linked to 9,20,21
Communication flow 12 and 1 that can be combined with 6 and 17
The user side closed networks 31, 32 and 33 are logically separated from each other on the network by 3, 14 and an independent server service is provided to each user side closed network 31, 32 and 33. Hereinafter, an embodiment will be described in which the hosts a, b, and c of the user-side closed networks 31, 32, and 33 use the server device 11 installed in the server-side closed network 10.

FIG. 3 shows V for identifying a closed network in the present invention.
It is an operation explanatory view of an inter-network communication method when a LAN tag is used. (Communication preparation) In each closed network 31, 32, 33,
It is assumed that the local addresses are freely used and assigned to the hosts a, b, and c, and the existence of other closed networks is unknown. The host a belongs to the closed network 31 and is assigned an IP address La (192.168.1.10). The host b belongs to the closed network 32 and has an IP address Lb (192.
168.1.10) has been assigned. Further, the host c belongs to the closed network 33 and has an IP address Lc (192.168.1.
10) is assigned. Since there is a possibility that the local address spaces will collide between the closed networks 31, 32 and 33, they are intentionally collided in this embodiment.

The access server (S) 26 of the server-side closed network 10 and the access servers (A) 28, (B) 29, (C) 30 of the user-side closed networks 31, 32, 33 include:
Gs (129.6
0.1.1), Ga (129.60.2.1), Gb (129.60.3.1), Gc
(129.60.4.1) is assigned.

The user-side closed network 31 has a local address Sa (192.168.1.1) for the server device 11, the user-side closed network 32 has Sb (192.168.1.1), and the user-side closed network 33 has S.
I am trying to allocate c (192.168.1.1). When the user-side closed networks 31 to 33 freely allocate the local address space to the server addresses, there is a possibility that the server addresses will collide with each other in the closed networks. Therefore, the collision is intentionally made in this embodiment. In accordance with this request, on the server device 11 side, in the communication flow 13 for allocating the server address Sa (192.168.1.1) for the closed network 32 to the logical device interface 23 (dev0) of the communication flow 12 for allocating the closed network 31 For the logical device interface 24 (dev1), the server address Sb (19
The server address Sc (192.168.1.1) is allocated to the logical device interface 25 (dev2) of the communication flow 14 in which 2.168.1.1) is allocated for the closed network 33.

At this time, in the lower mapping module 22 shown in FIG. 3, the logical device interfaces 23, 2
4, 25 and the protocol stacks 19, 20, 21 for managing the lower communication flow, and the protocol stacks 19,
Logical device interface 2 corresponding to 20, 21
At the same time as linking with 3, 24, and 25, the protocol stacks 19, 20,
21 (a pair of logical device interfaces 23, 24, 25) and a set of VLAN (Virtual LAN) tags 34, 35, 36 which are alternatives to a closed network using this lower communication flow [(inet0, dev0) ⇔ vlan60
0], [(inet1, dev1) ⇔vlan601], [(inet2, dev
2) ⇔ vlan602] and store it in the table. That is,
The values of the VLAN tags 34, 35, 36 identify closed networks. Since the conventional server device has only one internet protocol stack, it is impossible to assign the same server address to a plurality of logical device interfaces. However, in the present invention, a plurality of protocol stacks are provided. , Makes this possible.

Next, each user side closed network 31, 32, 33
Consider using the server device 11 as a Web server. The server device 11 has server processes 15 and 1
Which closed network 6 and 17 are used is recognized by the group ID of the process. A group ID (gid1000) is assigned to the user-side closed network 31, a group ID (gid1001) is assigned to the user-side closed network 32, and a group ID (gid1002) is assigned to the user-side closed network 33.
The server process of the Web server used by 3 is executed with the group ID assigned to each. At this time, the upper level mapping module 18 shown in FIG. 3 manages the upper level communication flow between the server process groups 15, 16, 17 and the protocol stacks 19, 20, 21 and corresponds to the server processes 15, 16, 17. At the same time as linking to the protocol stacks 19, 20, and 21, the protocol stack and group ID of each upper communication flow [inet0⇔gid1000], [inet1⇔g
id1001], [inet2⇔gid1002] are stored in the table.

As described above, the communication flows 12, 13, which are made by combining the upper communication flow and the lower communication flow,
Preparation of 14 is completed for each user side closed network 31, 32, 33. In the following, each user side closed network 31, 32,
A method in which the hosts a, b, and c of 33 communicate with the server device 11 installed in the server-side closed network 10 will be described.

(Communication Operation) (From Host of Closed Network on User Side to Server Device) Each host a, b, c belonging to the closed networks 31, 32, 33 of user side communicates with the server device 11 as a Web server. Think First, the host a is a local DNS (Domain Name System) built in the closed network 31.
server device 11 belonging to the server-side closed network 10 by m)
To obtain the local address Sa. Then, the packet is transmitted to the access server (A) 28. Similarly, the other hosts b and c have access servers (B) 2 respectively.
9, the packet is transmitted to (C) 30.

The access server (A) 28 which has received the packet from the host a performs an IP tunneling process.
That is, the IP packet header including the IP address La of the host a and the server address Sa is encapsulated with the global addresses of the access servers (A) 28 and (S) 26, and the access server (S) 26 is accessed via the Internet. Send the packet to. The same processing is performed in the other access servers (B) 29 and (C) 30, and each transmits the tunneled packet to the access server (S) 26.

When the access server (S) 26 receives the packet from the access server (A) 28 and uncapsulates it, the access server (S) 26 can identify the closed network 31 which is the transmission source of the packet in the communication thereafter. Tunnel and V
A LAN tag pair [(Ga, Gs) ⇔ vlan600] is stored in the table, and the corresponding tag (vlan600) is added to the packet. Then, according to the decapsulated IP header, the packet is transmitted to the server device 11. At this time,
The same applies to communication from the host b belonging to the closed network 32 and the host c belonging to the closed network 33. Each IP tunnel and VLAN tag pair [(Gb, Gs) ⇔vlan601],
According to the table that stores [(Gc, Gs) ⇔ vlan602],
VLAN tags (vlan601) and (vlan602) are added to the packet, respectively.

In the server device 11 which has received the tagged packet from the access server (S) 26, the VLAN
A pair (inet0, dev) of the protocol stack 19 and the logical device interface 23 matching the tag (vlan600)
0) is selected from the table of the lower mapping module 22, and the packet is transferred to the corresponding communication flow 12.
In the communication flow 12, packets are transferred from the logical device interface (dev0) 23 to the protocol stack (in
The lower communication flow up to et0) 19 is used to reach the server process (gid1000) 15 of the Web server linked to the protocol stack 19 using the upper communication flow. At this time, communications from the host b and the host c are also processed by the same process.

As described above, for the server device 11, La, L
Even if the sender's IP address collides like b and Lc,
Further, even if the server addresses assigned by the user-side closed networks 31, 32, and 33 collide with each other like Sa, Sb, and Sc, the communication flows 12 and 1 inside the server device 11 will be described.
3, 14 are logically separated on the network, and the server service can be provided completely independently of each closed network 31, 32, 33.

(Communication Operation) (From Server to Host on User-side Closed Network) Next, W
Consider that the eb server sends a reply packet to host a. Web server server process (gid1000) 15
Is the protocol stack (inet
0) A packet is returned to 19 using the upper communication flow. After packet processing is performed in each protocol of the protocol stack 19, a table [(inet0, d
ev0) ⇔ vlan600] according to the VLAN (vl
an600) is tagged, and from the logical device interface (dev0) 23 of the lower communication flow to the access server (S)
The packet is transmitted toward 26. At this time, reply packets to the hosts b and c are also processed by exactly the same process.

The access server (S) 26, which has received the tagged packet from the server device 11, carries out the IP tunneling process according to the table storing the pair of the IP tunnel and the VLAN tag [(Ga, Gs) ⇔vlan600]. That is, the server address Sa and the IP address L of the host a
Access server (S) for IP packet header including a
26, (A) 28 is encapsulated with a global address, and the access server (A) 2 is accessed via the Internet.
8 send packet. At this time, the reply packets to the hosts b and c are exactly the same, and the IP tunnel and V
LAN tag pair [(Gb, Gs) ⇔vlan601], [(Gc, G
s) ⇔ vlan602], access server (B) 29,
(C) Each packet is transmitted to 30.

Upon receiving the packet from the access server (S) 26, the access server (A) 28, when decapsulating, sends the packet to the host a according to the IP header after decapsulation. Finally, the host a receives the packet. Similarly, host b is access server (B)
From 29, host c receives the packet from access server (C) 30.

(Service Deployment for New User-side Closed Network) The user-side closed network (A) 31, which has already been registered,
When the user-side closed network (D) is newly added and registered with respect to (B) 32 and (C) 33, the server-side closed network 10 handles as follows. Select one from the unused communication flows, (1) Protocol stack: (inet3) (2) Group ID: (gid1003) (3) Upper mapping module: [inet3⇔gid1003] (4) Lower mapping module: ((Inet3, dev3) ⇔vl
an603] (5) Access server (S): By performing [(Gd, Gs) ⇔ vlan603], it is easy to create a new user-side closed network (D) without changing the environment of other closed networks. Virtual server services can be deployed.

(Server Device) FIG. 4 is a block diagram of a server device showing an embodiment of the present invention. As shown in FIG.
The server device 11 of the present invention includes a server process 41 that processes a plurality of processes, an upper mapping module 42 that manages an upper communication flow from the server process group to the protocol stack multiplex configuration unit 46, and a protocol stack multiplex configuration unit 46. It is composed of a lower mapping module 47 that manages a lower communication flow up to the physical network device multiplex configuration unit 51. Each of the upper mapping module 42 and the lower mapping module 47 is composed of combination storage units 44, 49 and link control units 43, 45 and 48, 50 connected to them.

The server device 11 is equipped with a plurality of communication flows each consisting of a combination of the upper communication flow from the server process 41 to the protocol stack 46 and the lower communication flow from the protocol stack 46 to the physical network interface or the logical device interface 51. . 1 these communication flows for each closed network on the user side
By allocating each of them, the individual communication paths in the server between each user-side closed network and the server device are logically separated on the network, and the user-side closed networks that are connected to each other are identified and separated. There is.

The server device 11 uses the unique communication flow assigned to each user-side closed network to hold the server address carried by the protocol stack multiplex configuration unit 46, control the packet delivery route, process each protocol packet, etc. Internet communication function, Web server, FTP (F
A server process such as an ile Transfer Protocol (server) is provided independently, and the user side closed network does not share the communication flow with other user side closed networks, and logically maintains the closedness on the network. As it is, it is possible to communicate with the server device 11.

As described above, the server process 41 is physically one server device, but the server process 41 is the link control unit 45.
Under the control of a plurality of server processes, and under the control of the link control unit 43, the protocol stack multi-tiered configuration unit 46 constitutes a plurality of protocol stacks. The upper level mapping module 42 stores the combination of the server process group and the protocol stack, and manages the upper level communication flow by link controlling the server process to the corresponding protocol stack. In addition, the lower mapping module 47
Multiplexing unit 51 of physical network interface or logical device interface and protocol stack 4
The lower communication flow is managed by storing the combination with 6 and link controlling the protocol stack to the corresponding physical network interface or logical device interface. Further, by providing a plurality of communication flows each consisting of a combination of the upper communication flow and the lower communication flow, and allocating one of these communication flows to each user-side closed network, the individual communication between each user-side closed network and the server device can be performed. The communication paths of are logically separated on the network.

(Server Device Program) The operation of the server device described in FIG. 4 and paragraph numbers [0041] to [0044] may be converted into a program, and the converted program may be recorded in a recording medium such as a CD-ROM. For example, the present invention can be easily realized by mounting this recording medium on a computer of an arbitrary server device in the server-side closed network and installing and executing the program.

[0046]

As described above, according to the present invention,
In network communication in which a plurality of closed networks on the user side are connected to the closed network on the server side by an access server using an IP tunnel, without sharing between the server function provided by the server device, the server process, and each closed network on the user side. , Each user side closed network can be provided independently. As a result, it becomes possible to physically operate one server device installed in the server-side closed network as if it were a dedicated server belonging to each user-side closed network. As a result, a housing service that provides the construction and operation of a shared server for multiple closed networks, which is required during the implementation of projects that require data exchange between multiple companies, which is expected to increase in the future, and the implementation of procurement operations. Can be expanded to.

Since this service enables information to be linked with a plurality of other companies without modifying the closed network of the company, it can be deployed as a new additional service of the VPN service provided by ISPs and the like. Can be expected. Furthermore, it will be possible to expand to portal site services for individuals as well as companies. Further, in the present invention, an administrator of each user closed network can freely assign an address to one server device and can receive a push-pull type service. As described above, the present invention has a remarkable effect that it can be used as a means for constructing a new information distribution platform for closed networks.

[Brief description of drawings]

FIG. 1 is a network configuration diagram to which an inter-network communication method of the present invention is applied.

FIG. 2 is a diagram showing a conventional Internet protocol stack.

FIG. 3 is an operation explanatory diagram of an inter-network communication method showing an embodiment of the present invention.

FIG. 4 is a configuration diagram of a server device showing an embodiment of the present invention.

[Explanation of symbols]

10 ... Server-side closed network, 11 ... Server device, 12 ... Communication flow (for closed network A), 13 ... Communication flow (closed network B)
, 14 ... Communication flow (for closed network C), 15, 16,
17 ... Server process group, 18 ... Upper mapping module, 19, 20, 21 ... TCP / IP protocol group,
22 ... Lower mapping module, 23, 24, 25 ...
Logical device interface, 26 ... Access server (S), 27 ... Internet, 28, 29, 30 ... Access server (A), (B), (C), a, b, c ... Host, 3
1, 32, 33 ... Closed networks (A), (B), (C), 3
4, 35, 36 ... VLAN tag, 41 ... Server process, 42 ... Upper mapping module, 43, 45 ... Link control unit, 44 ... Combination storage unit, 46 ... Protocol stack multiplexing configuration unit, 47 ... Lower mapping module, 48, 50 ... Link control unit, 49 ... Combination storage unit, 51 ... Logical network device multiplex configuration unit.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5B089 GB01 HB02 HB11 KB06 KF06                       KG08                 5K033 AA04 CB08 DA05 DB12 DB14

Claims (7)

[Claims] 1. An inter-network communication method using a single server device physically in a server-side closed network in which a plurality of user-side closed networks and IP tunnels are used for connection by an access server. A higher communication flow managed by a module that stores a combination of a process group and a protocol stack and that controls a link between the server process group and the protocol stack according to the stored content, a physical network interface or a logical device interface, and the protocol. A plurality of communication flows that are stored in combination with a stack and are combined with a lower communication flow managed by a module that links the protocol stack to a corresponding physical network interface or logical device interface, By allocating one communication flow to each closed network on the user side, individual communication paths in the server device between each closed network on the user side and the server device are logically separated on the network. A method of network communication between a closed network of multiple users and a closed network of servers. 2. The network communication method according to claim 1, wherein in communication with each user-side closed network connected to the physical network interface or logical device interface for each communication flow via the network, the protocol of the communication flow is used. A network communication method between a closed network on the side of a plurality of users and a closed network on the side of a server, wherein an independent server service is provided to each closed network on the side of a user by using a stack and a group of server processes. 3. The network communication method according to claim 1, wherein an IP is applied to a physical network interface or a logical device interface for each of the plurality of communication flows.
A network communication method between a closed network on the side of multiple users and a closed network on the side of a server, which allows overlapping addresses to be assigned. 4. A physically single server device provided in a server-side closed network that is connected by an access server using a plurality of user-side closed networks and IP tunnels, and comprises a server process group and a protocol stack. A means for managing the higher-level communication flow by storing the combination and controlling the link between the server process group and the protocol stack according to the stored contents, and a combination of the physical network interface or logical device interface and the protocol stack. And a communication flow composed of a combination of the upper communication flow and the lower communication flow, by storing the above information and linking the protocol stack to a corresponding physical network interface or logical device interface. By equipping a plurality of units and allocating one communication flow to each closed network on the user side, individual communication paths in the server device between each closed network on the user side and the server device are logically separated on the network. A server device comprising: a communication unit. 5. The server apparatus according to claim 4, wherein in communication with each user-side closed network that connects to a physical network interface or a logical device interface for each communication flow via the network, the protocol of the communication flow is used. A server device comprising: a virtual server providing means for performing an independent server service for each user-side closed network by using a stack and a group of server processes. 6. A server device program for causing a computer to realize the operation of the server device according to claim 4. 7. A computer-readable recording medium having the server device program according to claim 6 stored therein.